Ординатура / Офтальмология / Английские материалы / Color Atlas of Herpetic Eye Disease A Practical Guide to Clinical Management_Sundmacher_2009

than currently recognizable. However, as these late dendritic efflorescences invariably heal with consequent dry eye therapy only, their clinical character is clearly trophic. Reports that topical steroids are sometimes needed to successfully treat such conditions are in accordance with my own positive experiences with Cs-A eye drops in recalcitrant severe filiform keratitis: By symptomatically reducing the grade of inflammation in the conjunctiva, the pathologic conjunctival mucous production is reduced. Mucous production seems to be the most important cause for the late pseudodendritic efflorescences. Consequent therapy with mucosolvents, sufficient tear substitution without preservatives and potentially in addition a cautious anti-inflammatory therapy with minimally dosed steroids or potent non-steroidal anti-in- flammatory substances, e. g., Cs-A eye drops, will probably be a good combination to get rid of these chronic epithelial efflorescences after zoster.

References

Al-Muammar A, Jackson WB (2004) Management of ophthalmic zoster mucous plaques keratopathy: report of three cases. Can J Ophthalmol 39: 74–76

Chen HJ, Pires RT, Tseng SC (2000) Amniotic membrane trans- plantation for severe neurotrophic ulcers. Br J Ophthalmol 84:

826–833

Cobo LM (1988) Corneal complications of herpes zoster ophthalmicus. Prevention and treatment. Cornea 7: 50–56

Engstrom RE, Holland GN (1988) Chronic herpes zoster virus kera- titis associated with the acquired immunodeficiency syndrome. Am J Ophthalmol 105: 556–558

Late Corneal Complications after Zoster

Hedges TR, Albert DM (1982) The progression of the ocular abnor- malities of herpes zoster: histopathologic observations of nine cases. Ophthalmology 89: 165–176

Kaufman SC (2008) Anterior segment complications of herpes zoster ophthalmicus. Ophthalmology 115 Suppl: S24–S32

Liesegang TJ (1985) Corneal complications of herpes zoster ophthal- micus. In: PC Maudgal, L Missotten (eds) Herpetic Eye Diseases. Doc Ophth Proc Ser 44, pp 395–402

Marsh RJ (1973) Herpes zoster keratitis. Trans Ophthalmol Soc

UK: 93: 181–192

Marsh RJ, Cooper M (1987) Ophthalmic zoster: mucous plaque ker- atitis. Br J Ophthalmol 71: 725–728

Mondino BJ, Brown SI, Mondzelewski JP (1978) Peripheral corne- al ulcers with herpes zoster ophthalmicus. Am J Ophthalmol 86: 611–614

Pavan-Langston D, Yamamoto S, Dunkel EC (1995) Delayed her- pes zoster pseudodendrites. Polymerase chain reaction detection and a role for antiviral therapy. Arch Ophthalmol 113: 1381–1385

Piebenga LW, Laibson PR (1973) Dendritic lesions in herpes zoster ophthalmicus. Arch Ophthalmol 90: 268–270

Waring GO, Ekins MB (1981) Corneal perforation in herpes zoster ophthalmicus caused by eyelid scarring with exposure kerati- tis. In: R Sundmacher (ed) Herpetic Eye Diseases. Bergmann,

Munich, pp 469–478

Womack LW, Liesegang TJ (1983) Complications of herpes zoster ophthalmicus. Arch Ophthalmol 101: 42–45

Zaal MJW, Maudgal PC, Rietveld E, Suir P (1991) Chronic ocular zoster. Curr Eye Res 10 (Suppl): S125–130

Keratoplasty

in Zoster Eyes

Core Messages

•Different to keratoplasty in HSV eyes, recurrences of VZV disease in the graft are normally not to be expected, and a systemic antiviral prophylaxis is therefore unnecessary.

•Because trophic problems cause a good deal of complications after keratoplasty in zoster eyes, if possible, it is not recommended to perform systemic immune prophylaxis as usual with steroids, but with non-steroidal agents, above all with cyclosporine A or mycophenolatemofetile.

Many common aspects, problems, complications, and available solutions for perforating keratoplasty have already been presented in detail in Sect. 2.14 on surgery in HSV eyes. In this section, emphasis can be laid, therefore, on the major differ- ences between keratoplasty in VZV disease as compared with keratoplasty in HSV disease:

•The immunological problems and the risk of immune reactions and immunologic graft failure are identical in both groups, as are the chances and indications for using HLA matched donor material.

•The risk of recurring viral disease is high in HSV infection, and it is normally zero in zoster eyes. Except for the special situation in severely immune compromised individuals, and except for the rare event of zoster in children or young adults, which make a second bout of zoster theoretically possible after many decades, zoster does not recur a second time in a patient. Therefore, antiviral prophylaxis is unnecessary. This is an important advantage of keratoplasty in zoster eyes.

•This advantage is counteracted, however, by much more trophic problems after severe zoster than after recurrent HSV disease. These trophic disturbances often persist in the time period after keratoplasty, and they may dangerously interfere with the integrity and function of the graft. A special danger comes from the circumstance where severe trophic disease is mostly incompatible with high-dosed steroid prophylaxis of immune reactions, especially with topical ste-

Section

3.9

roids. If the surgeon refrains from steroids in an attempt not to allow trophic disease to establish, the battle for survival of the corneal graft has already been lost in many cases. The solution of this problem is to administer non-steroidal systemic immunosuppressive agents, above all cyclosporine A or mycophenolate mofetil, for the time period of the highest immune threat, i. e., for 1–1.5 years. If this is done not only for immunologically especially compromised zoster cases but for all zoster keratoplasties (!), and if simultaneously the risk of trophic disturbances is consequently reduced by efficient dry eye prophylaxis, then the prognosis of keratoplasty in zoster eyes may approach the excellent prognosis of keratoplasty in HSV eyes.

•In contrast to severe HSV disease, where keratoplasty à chaud can often shorten the course of disease and reduce the overall need for systemic therapy, costs included, ker- atoplasty à chaud in zoster eyes can not be recommended.

Severe trophic problems are the most frequent reason for emergency situations in zoster, and these severe trophic problems continue to be operative also after keratoplasty

à chaud. If it is possible to defer keratoplasty up to a time when a sufficient percentage of the trophic problems has resolved or at least ameliorated, the chances for successful graft survival become much better. It is always worthwhile, therefore, to try to defer keratoplasty in zoster eyes until the eye has become clinically quiet.

Analysis of Fig 3.39

The zoster patient in Fig. 3.39 was treated at a time, when non-ste- roidal immune suppression was not yet available for keratoplasty. Thus, he was given systemic and topical steroids “as usual” together with dry eye prophylaxis. When he returned to his surgeon for a control 6 weeks after surgery, the lower half of the cornea had become dramatically opaque in a half-moon shaped area. The graft epithelium had fallen off and no host epithelium had come in to cover the defect (Fig. 3.39a). The Bengal rose stain (Fig. 3.39b) discloses that the area of diseased epithelium is even larger than the epithelial ulcer. As no immune reaction was detectable and the miserable state of the graft had exclusively been attributed to trophic reasons, it was decided to carry on for some months with a moderate systemic steroid prophylaxis, to refrain of all topical steroids, and, instead, to enhance the topical dry eye therapy. The success of these simple measures was stunning. 12 weeks later (Fig.3.39c), the defect had closed. The pupillary area had become

much more transparent, and only a dense superficial calcification from the intensified application of phosphate-containing eye drops was a new unpleasant complication. 1.5 years after keratoplasty, the graft had functionally again become crystal clear in the center. The calcifications had spontaneously been resorbed after reduction of artificial eye drop application from every 0.5 hr to 7 qd. Immune reactions had not interfered, and new trophic problems did not interfere either.

Analysis of Fig 3.40

The treatment of the 77-year-old zoster patient in Fig. 3.40 also dates back to the time when neither ACV nor non-steroidal systemic immune-suppressive agents were available. When he showed-up first 10 days after beginning of zoster, the cornea was already superficially ulcerated, and the ulceration quickly proceeded to comprise nearly the whole inferior cornea. The lids did not close tightly, and the ulceration was judged to be enforced by intermittent lagophthalmus, i. e., by exposure. In spite of antibiotic therapy, a hypopyon developed (Fig. 3.40a). Shortly after, a paralimbal pinpoint perforation was detected (Fig. 3.40b), and the cornea was covered with an oxygen-permeable silicon contact lens, which was then “en vogue” for sealing of corneal perforations. The perforation broadened, however, under the silicon

lens, and emergency surgery became inevitable (Fig. 3.40c). The right time for a conjunctival or full corneal cover had long been gone, and only a perforating keratoplasty procedure could help in this situation. The limbal position of the perforation and the size of the ulcer would have made the use of an oversized sclerocorneal graft necessary with the great danger of immunologic failure. In an attempt to at least save the still clear upper half of the patients cornea, only the lower half was resected “free hand” and the defect filled with a full thickness corneal graft (Fig. 3.40d). The further course was unexpectedly positive. Half a year later, the donor as well as the host cornea were both clear and the eye was clinically quiet (Fig. 3.40e). The mature cataract, which had developed in the course of perforation and keratoplasty, was secondarily successfully operated, and the patient gained reasonable vision with this previously almost lost eye without further surgical interventions.

Retrospectively, it was waited too long for the ulcer to heal with conservative measures only. A quicker emergency surgery with amnion, conjunctiva, or cornea to cover the large ulcer would have been less risky and would probably have spared some emotional stress of the patient and the surgeon. The unexpectedly good outcome in this case should not lead to the erroneous assumption that keratoplasty á chaud has a mostly good prognosis in ophthalmic zoster.

Analysis of Fig 3.41

Figure 3.41 shows how the case in Fig. 3.40 could have developed if the large basal ulcer would have been covered in time by a conjunctival flap. After an adequate consolidation time of 6–12 months, such an eye can safely be operated, especially if potent non-steroidal systemic immune-suppressive agents are available. The massive vascularization is no longer a major problem with substances like cyclosporine A or mycophenolate mofetil. The risk of immune reaction can be further reduced, and the chances for a clear long-term survival of the corneal graft can be increased by use of a well HLA-matched corneal transplant.

Analysis of Fig 3.42

The case in Fig. 3.42 is rather special, but worthwhile demonstrating for principal reasons. It is common knowledge that keratoplasty wounds never reach the tensile strength of normal corneal tissue and that the life-long danger of wound rupture after blunt ocular trauma exists for every keratoplasty eye. However, this danger may be especially great in zoster eyes.

In the 1960s, a surgical treatment of zoster keratitis was occasionally tried and it was called “circumcision”. The idea was to circumferentially cut all sensory nerves and thus block VZV from invading the cornea centrally to the circumcision line. Basically, simple

trephination of the cornea down to the level of Descemets was performed, and sometimes Descemet´s was certainly perforated on this occasion, so that the resultant state was similar to that after perforating keratoplasty. The 90-year-old lady presented 22 years after this procedure with a widely lysed corneal wound. She denied any trauma. 2 weeks previously, a corneoscleral cataract extraction had been performed. It seems then, that the minor trauma of lifting the corneoscleral flap in the course of cataract surgery was sufficient to open a “keratoplasty” wound in a zoster eye after 22 years. Such experience may be worthwhile to note in order to teach especially zoster patients with a corneal transplant to avoid even minor ocular trauma.

References

Alino AM, Perry HD, Kanellopoulos AJ, Donnenfeld ED, Rahn EK (1998) Conjunctival flaps. Ophthalmology 105: 1120–1123

Mackensen G, Sundmacher R, Witschel H (1985) Late wound com- plications after circular keratotomy for zoster keratitis. Cornea 3: 95–98

Marsh RJ, Cooper M (1989) Ocular surgery in ophthalmic zoster. Eye 3: 313–317

Reed JW, Joyner SJ, Knauer WJ (1989) Penetrating keratoplasty for herpes zoster keratopathy. Am J Ophthalmol 107: 257–261

Soong HK, Schwartz AE, Meyer RF, Sugar A (1989) Penetrating keratoplasty for corneal scarring due to herpes zoster ophthal- micus. Br J Ophthalmol 73: 19–21

Tanure MA, Cohen EJ, Grewal S, Rapuano CJ, Laibson PR (2000)

Penetrating keratoplasty for varicella-zoster virus keratopathy. Cornea 19: 135–139

Van Gelderen BE, van der Lelij A, Treffers WF, van der Gaag R

(2000) Detection of herpes simplex virus type 1, 2 and varicella zoster virus DNA in recipient corneal buttons. Br J Ophthalmol

84: 1238–1243

The basic similarities (see Sect. 3.1) and the many dissimilarities between VZV and HSV ocular disease have already been discussed or shortly been mentioned in the preceding sections on HSV and VZV diseases. Therefore, for this concluding section it may suffice to compile the various aspects in an ab-

breviated form in Table 3.2. Some statements in this table do not fulfil the requirements of unequivocal scientific evidence.

Instead, they express the author´s belief on how to fit together clinical observations with a currently still partially hypothetical pathophysiological background.

Regularly indicated even at higher doses together with efficient antiviral cover

References

Ahonen R, Vannas A (1985) Clinical comparison between herpes simplex and herpes zoster ocular infections. In: PC Maudgal, L Missotten (eds.) Herpetic Eye Diseases. Doc Ophthal Proc Ser 44, pp 389–394

Beards G, Graham C, Pillay D (1998) Investigation of vesicular rash- es for HSV and VZV by PCR. J Med Virol 54: 155–157

Croen KD, Ostrove JM, Dragovic LJ, Straus SE (1988) Patterns of gene expression and site of latency in human trigeminal ganglia are different for varicella-zoster and herpes simplex virus. Proc. Natl Sci USA 85: 9773–9777

Forrest WM, Kaufman HE (1976) Zosteriform herpes simplex. Am

J Ophthalmol 81: 86–88

157

Kalman CM, Laskin OL (1986) Herpes zoster and zosteriform herpes simplex virus infections in immunocompetent adults. Am J Med 81: 775–778

Rubben A, Baron JM, Grussendorf-Conen EL (1997) Routine de- tection of herpes simplex virus and varicella zoster virus by polymerase chain reaction reveals that initial herpes zoster is frequently misdiagnosed as herpes simplex. Br J Dermatol 137:

259–266

Straus SE, Meier JL (1992) Comparative biology of latent varicel- la-zoster virus and herpes simplex infections. J Infect Dis 166 (Suppl 1): S13–S23

Thygeson P, Ostler HB (1979) Zoster and herpes simplex virus uveitis: a comparison. In: AM Silverstein, GR O`Connor (eds) Immunology and Immunopathology of the Eye. Masson, USA, pp 230–240

Cytomegalovirus (CMV)

Diseases of the

Anterior Segment

Rainer Sundmacher and Johannes Stammen

Core Messages

•A connatal and two adult cases of monosymptomatic endotheliitis plus trabeculitis could be correlated with culturable CMV or the presence of CMV DNA in the aqueous humor.

•Cases with Posner–Schlossman syndrome are prime suspects for CMV disease of the anterior segment of the eye and should be accordingly investigated.

Intrauterine transmitted connatal viral ocular disease has been omitted as a topic for this book. The presentation and discussion would have had to include too many special fields of virology, obstetrics, and pediatrics, which are outside the scope of this atlas. Several intrauterine acquired virus infections may cause malformations and a variety of diseases. Sometimes also the eye becomes affected, e.g., the lens and also the cornea in the course of systemic rubella or CMV infection. The viral etiology can then mostly be easily suspected from the signs of general viral disease. Laboratory confirmation is also mostly not a problem. These polysymptomatic eye affections in the course of systemic intrauterine viral disease will not be further discussed here.

However, a special monosymptomatic connatal CMV disease of the eye will be presented because it fits well with the finding of monosymptomatic endotheliitis of suspected CMV etiology in adults. It seems that we are now able to define much more precisely as ever which type of inflammation of unknown origin in the anterior segment of the eye may be caused by CMV.

Analysis of Fig 4.1

The newborn in Fig. 4.1 presented with a bilateral connatal inflammatory glaucoma with both corneae enlarged, but without the typical signs of idiopathic buphthalmus. There were no Haab lines, and the corneas were too opaque to disclose details in the eye. Intraocular pressure was 27 mm Hg RE and 50 mm LE. The suspected diagnosis was infectious endotheliitis and trabeculitis, with herpes viruses ranging among the most suspect viruses. CMV

Chapter

4

was grown from aqueous taps from both eyes on three occasions, the last time at the age of 10 months (Fig. 4.1c). Up to this time, all therapeutic attempts including trabeculotomies, medical glaucoma treatment, and also an antiviral trial with ARA-A, which was the only available experimental systemic antiviral agent in those years, had failed. The intraocular pressure was still high with the corneas further enlarged and scarred. There was no other help. The child remained blind bilaterally.

The intensive search for associated general viral disease remained constantly negative. The child was perfectly healthy except for the bilateral eye affection, and the child also developed somatically and mentally normal except for the incurable connatal blindness. The anterior chamber was not the only site where CMV was isolated from after birth. Virus was also repeatedly grown from conjunctival swaps and from the urine, though not regularly. A general CMV disease, however, was not present.

Therefore, this case was classified as the presumably first description of a monosymptomatic severe CMV endotheliitis and trabeculitis, with the infection acquired intrauterine, presumably rather late in the course of gravidity. The mother had never shown signs of clinical viral disease. However, she had high IgM-titers against CMV after birth.

The publication of this case in the proceedings of the first international herpes symposium certainly did not promote its knowledge.

Recently, a small series of observations has been published of CMV DNA findings in the aqueous humor of adults with endotheliitis and uveitis of unknown origin. Elevated intraocular pressure was also involved. Therefore, it becomes probable that monosymptomatic corneal endotheliitis associated with elevated pressure is a common feature of CMV infection of the anterior segment of the eye.

The clinical picture of connatal and adult diseases are quite different as far as acuity and severeness are concerned.

Pathophysiologically, however, they follow an identical infection scheme. As endotheliitis with elevated pressure is still a somewhat imprecise description, which covers several viral etiologies and possibly also non-viral etiologies, the question is, whether additional signs exist which allow to more precisely predict which type of endotheliitis is a candidate for CMV eti- ology in adults. Two of my own cases of CMV DNA finding in monosymptomatic endotheliitis both fulfil the diagnostic criteria of Posner–Schlossman syndrome.

Analysis of Fig 4.2

The 67-year-old male patient in Fig. 4.2 first presented with a 33year long history of “recurrent iritis with increasingly severe secondary glaucoma, while the iritis was always stunningly mild with only very sparse precipitates”.

This is a usual description of Posner–Schlossman syndrome with the typical course of glaucoma. In the first years or decades, glaucoma is only transient and quickly fades away with and without therapy. With symptomatic therapy (topical steroids), the attacks seem to have a shorter duration, which has led to the widespread conviction that some etiologically unclear acute “allergic” swelling of trabecular meshwork cells is the cause of disease, i. e., an immunologic trabeculitis of unknown origin.

We and others have favored the hypothesis of a viral etiology, especially of members of the herpes family. As HSV and VZV are typically causing endotheliitis and trabeculitis, they were prime candidates for etiology. HSV has been excluded. An etiologic role for VZV could never be unequivocally proven. It was some surprise when CMV PCR was positive in the aqueous humor of this patient on occasion of trabeculectomy, which became necessary, because the secondary glaucoma no longer responded to symptomatic therapy.

This is the second case after the newborn patient (Fig. 4.1) in which we could correlate trabeculitis and endotheliitis with a positive CMV finding in the aqueous humor.

Analysis of Fig 4.3

The third case of endotheliitis and trabeculitis with a positive CMV PCR in the aqueous humor was found in a 30-year-old male who had suffered from yearly attacks of Posner–Schlossman syndrome for three years (Fig. 4.3). The disease had been classified as “some kind of atypical herpes disease with endotheliitis and glaucoma” and had already been treated as such with ACV plus steroids.

The complex history of these two adult patients is reserved for a separate publication after more follow-up time. For this atlas, it shall suffice to document the finding of intraocular CMV in two adults with endotheliitis and trabeculitis, and to elaborate a bit more on the clinical signs of this disease.

Posner–Schlossman syndrome is probably no clear-cut entity, i. e., it has more than one cause. This can be suspected from the observation that number, distribution, fine structure, and persistence of the endotheliitic precipitates are quite variable in Posner–Schlossman syndrome. The common link of all Posner–Schlossman syndromes, however, is the correlation of very subtle signs of endotheliitis – as evidenced by the typical endotheliitic precipitates – with a disproportionate acute rise in intraocular pressure by presumed trabeculitis.

If all cases from the potentially heterogenous Posner– Schlossman group are systematically punctured, and the aqueous humor investigated by PCR for the presence of herpes